ai-safety-foundation / sparse_autoencoderLinks
Sparse Autoencoder for Mechanistic Interpretability
☆268Updated last year
Alternatives and similar repositories for sparse_autoencoder
Users that are interested in sparse_autoencoder are comparing it to the libraries listed below
Sorting:
- ☆348Updated last month
- Using sparse coding to find distributed representations used by neural networks.☆272Updated last year
- ☆242Updated last year
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆221Updated 9 months ago
- Mechanistic Interpretability Visualizations using React☆291Updated 9 months ago
- ☆127Updated this week
- Sparsify transformers with SAEs and transcoders☆630Updated last week
- ☆127Updated last year
- ☆174Updated 10 months ago
- ☆187Updated 2 months ago
- The nnsight package enables interpreting and manipulating the internals of deep learned models.☆674Updated last week
- Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models …☆215Updated last week
- Training Sparse Autoencoders on Language Models☆985Updated this week
- Steering vectors for transformer language models in Pytorch / Huggingface☆125Updated 7 months ago
- Steering Llama 2 with Contrastive Activation Addition☆187Updated last year
- ☆525Updated last year
- ☆54Updated 10 months ago
- ☆124Updated last year
- This repository collects all relevant resources about interpretability in LLMs☆373Updated 11 months ago
- Open source replication of Anthropic's Crosscoders for Model Diffing☆59Updated 11 months ago
- Sparse probing paper full code.☆61Updated last year
- Tools for understanding how transformer predictions are built layer-by-layer☆530Updated 2 months ago
- ViT Prisma is a mechanistic interpretability library for Vision and Video Transformers (ViTs).☆309Updated 2 months ago
- An Open Source Implementation of Anthropic's Paper: "Towards Monosemanticity: Decomposing Language Models with Dictionary Learning"☆49Updated last year
- Sparse Autoencoder Training Library☆54Updated 5 months ago
- Modified to support crosscoder training.☆23Updated 2 months ago
- For OpenMOSS Mechanistic Interpretability Team's Sparse Autoencoder (SAE) research.☆154Updated this week
- Attribution-based Parameter Decomposition☆31Updated 3 months ago
- Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".☆278Updated 3 months ago
- ☆199Updated 10 months ago